Track 3D AI - Construction Analytics Platform

AI-powered construction project analytics and forecasting platform.

Serverless Build: Uses Neon.tech, PlanetScale, and Supabase for production-grade scalability.

⚠️ Running on a free-tier Neon.tech server. The first API response may take up to 50 seconds or more to become active due to cold start delays.

Features

• AI-Powered Analytics: Predict project completion dates, resource needs, and deviations.

• Real-Time Progress Tracking: Compare planned vs. actual progress.

• Role-Based Access Control:

  • Core User: Access analytics and reports.

  • Admin User: Full project management control.

• Dashboards & Reports: Generate reports with real-time data insights.

Chart

Installation

Prerequisites

• Python 3.11+

• FastAPI & TypeScript

• Docker

• Amazon S3 Bucket

Setup

Clone the repository

git clone https://github.com/MisfiT2020/Track-3D.git
cd Track-3D

Install backend dependencies

pip install -r requirements.txt

Install frontend dependencies

cd frontend
npm install

Configuration

• Update the address in vite.config.ts as needed.

Frontend Environment Setup

• Create a .env file inside the frontend directory and set your API base URL:

VITE_API_BASE_URL= <replace with backend url>

Make sure the backend server is accessible from this URL in your deployment.

Branches / Builds

This project supports two types of database builds, each maintained under separate branches:

• Main Branch (main) – Uses a serverless/hosted database (e.g., Supabase, PlanetScale) for production-ready, scalable deployment.
• Static DB Branch (static-db) – Uses a local SQLite DB for quick setup, development, and offline demos.

Switch to the branch that suits your use case:

git checkout main        # For hosted serverless DB  
git checkout static-db   # For local static SQLite DB

Run the Application

Start the application using Docker

sudo docker compose up --build

Screenshots

ADMIN INTERFACE

USER INTERFACE

Usage

• Log in to access your projects.

• Upload project plans and monitor progress.

• View AI-generated insights and analytics.

• Generate and export reports.

Default Credentials (Admin Demo Account)

Username: misfit  
Password: misfit

Normally, when signing-up from site, requires normal password requirements. > but it's a demo. so, makin it easier to login.

Permissions

CLI Method

• To update a account as sudo, we need to need make the sudo account from swagger ui

SWAGGER Method

• To update from swagger ui, go to running backend port <url>:<port>/docs#/
• select sign-up field

{
  "username": "misfit",
  "email": "misfit",
  "password": "misfit",
  "is_sudo": true
}

🔹 change the is_sudo to change the admin permission for the user when creating